A Financial Model Comparing Car Ownership with UberX in Los Angeles

Uber is great, but the biggest cost is that you need to pay for the services of someone driving for you. I live in Montreal and regularly use http://communauto.com/ They have two services, 1) for ~$2,50/hour + $0,10/km, where you need to make a reservation, you get the car from one specific parking lot and have to return it there. It's useful only if you plan in advance, if you haven't reserved a car for the weekend by Wednesday forget about it.

The 2) service though you just swipe your metro card on the panel for an auto-mobile car parked somewhere, drive for $0,30/minute and leave the car anywhere (in some very big zones in the cities). It's exactly like https://www.car2go.com/ except their auto-mobile vehicles (hybrids and electrics) have backseat, so we can go around with kids (whereas you can't with a Smart).

I commute by bicycle to work, so we use this only on the weekends. It takes around $7 to go from our place to the center, while public transport would be $5, so it's really great for the convenience and speed.

By far it really beats owning a car in Montreal in cost and convenience. Parking space is limited and very expensive ($8 for 2 hours in city center). If you have a car you have also to keep changing parking even if you don't use it because there are very few buildings with garage and they clean the streets at least once or twice a week for the entire year. I've been car-free for over 20 years, it's a huge relief to not have to worry about taxes, maintenance, robbery, etc etc that comes with the responsibility of owning a car.

Come self-driving cars, we might have only some very few people that will keep owning their own vehicles due to some nostalgia feeling.

LA is also a very bipolar town, car-wise. If you have a 5+ mile commute, it's going to be way worse than the same distance in most US cities, because of the traffic. But the flip side is that there is far more density to the LA suburbs than most anywhere else in the country—it's hard to even use the same word to describe suburban LA (7K/sqmile average for the whole urbanized area per Wikipedia, with parts significantly higher) as, say, suburban DFW, where the population density is half that for Dallas proper alone, let alone all the suburbs.

So if you live near where you work, you can bike or walk or bus and there's a decent chance you'll be in walking or biking distance of some shops and restaurants as well. And then the occasional Uber or cab for longer trips (or just bumming off of friends who have cars) is extremely practical.

But if you don't telecommute, being able to live near where you work is something that's a lot easier for those without a family. And then those Ubers would add up fast!

(The other thing I've anecdotally noticed since moving to LA, is that the willingness to sit in traffic for an hour each way to commute seems way higher in LA natives than in transplants—and the reverse is true of the willingness to pay the premium to live close to work and avoid that.)

That is certainly the thesis behind the wild valuation of Uber - but so far Uber hasn't really shown the ability to lock customers into its taxi network. Whoever develops the self driving cars can build the Uber app/network part of the system much more easily than Uber will be able to build the self driving car part of the system.

And if that happens, there are plenty of car dealers ready to take your money. It only takes 2-3 days to buy a car, people do it all the time.

I grew up in LA - Uber is a practical car replacement option if you stay largely within your neighborhood or city.

This isn't how most people in the basin live however - many (if not most) live at least 10, but usually closer to 20-25 miles from work, and rely on the freeway to get them to work - 50+ miles a day even in an UberX is not practical from a cost basis - not when you consider the large number of people who don't go out and buy a large luxury car, but instead pick up a beater economy car from craigslist, or something that they bought new a decade or more prior.

The comparison however does work when you consider someone who commutes under 10 miles each way a day, and buys a new car every 3-5 years.